PraisonAI 0.0.57__tar.gz → 0.0.59rc2__tar.gz

{praisonai-0.0.57 → praisonai-0.0.59rc2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PraisonAI
-Version: 0.0.57
+Version: 0.0.59rc2
 Summary: PraisonAI application combines AutoGen and CrewAI or similar frameworks into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customization, and efficient human-agent collaboration.
 Author: Mervin Praison
 Requires-Python: >=3.10,<3.13
@@ -17,6 +17,7 @@ Provides-Extra: cohere
 Provides-Extra: google
 Provides-Extra: gradio
 Provides-Extra: openai
+Provides-Extra: train
 Provides-Extra: ui
 Requires-Dist: agentops (>=0.2.6) ; extra == "agentops"
 Requires-Dist: aiosqlite (>=0.20.0) ; extra == "chat" or extra == "code"

{praisonai-0.0.57 → praisonai-0.0.59rc2}/praisonai/cli.py

@@ -98,6 +98,12 @@ class PraisonAI:
         if getattr(args, 'code', False):
             self.create_code_interface()
             return
+
+        if args.agent_file == 'train':
+            from .train import main as train_main
+            train_args = sys.argv[2:]  # Get all arguments after 'train'
+            train_main(train_args)  # Pass the arguments to train.py's main function
+            return 
         
         invocation_cmd = "praisonai"
         version_string = f"PraisonAI version {__version__}"

{praisonai-0.0.57 → praisonai-0.0.59rc2}/praisonai/deploy.py

@@ -56,7 +56,7 @@ class CloudDeployer:
             file.write("FROM python:3.11-slim\n")
             file.write("WORKDIR /app\n")
             file.write("COPY . .\n")
-            file.write("RUN pip install flask praisonai==0.0.57 gunicorn markdown\n")
+            file.write("RUN pip install flask praisonai==0.0.59rc2 gunicorn markdown\n")
             file.write("EXPOSE 8080\n")
             file.write('CMD ["gunicorn", "-b", "0.0.0.0:8080", "api:app"]\n')
             

praisonai-0.0.59rc2/praisonai/train.py

@@ -0,0 +1,232 @@
+import subprocess
+import os
+import sys
+import yaml
+import torch
+from transformers import TextStreamer
+from unsloth import FastLanguageModel, is_bfloat16_supported
+from trl import SFTTrainer
+from transformers import TrainingArguments
+from datasets import load_dataset, concatenate_datasets, Dataset
+from psutil import virtual_memory
+
+class train:
+    def __init__(self, config_path="config.yaml"):
+        self.load_config(config_path)
+        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        self.model, self.tokenizer = None, None
+
+    def load_config(self, path):
+        with open(path, "r") as file:
+            self.config = yaml.safe_load(file)
+
+    def print_system_info(self):
+        print(f"PyTorch version: {torch.__version__}")
+        print(f"CUDA version: {torch.version.cuda}")
+        if torch.cuda.is_available():
+            device_capability = torch.cuda.get_device_capability()
+            print(f"CUDA Device Capability: {device_capability}")
+        else:
+            print("CUDA is not available")
+
+        python_version = sys.version
+        pip_version = subprocess.check_output(['pip', '--version']).decode().strip()
+        python_path = sys.executable
+        pip_path = subprocess.check_output(['which', 'pip']).decode().strip()
+        print(f"Python Version: {python_version}")
+        print(f"Pip Version: {pip_version}")
+        print(f"Python Path: {python_path}")
+        print(f"Pip Path: {pip_path}")
+
+    def check_gpu(self):
+        gpu_stats = torch.cuda.get_device_properties(0)
+        print(f"GPU = {gpu_stats.name}. Max memory = {round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3)} GB.")
+
+    def check_ram(self):
+        ram_gb = virtual_memory().total / 1e9
+        print('Your runtime has {:.1f} gigabytes of available RAM\n'.format(ram_gb))
+        if ram_gb < 20:
+            print('Not using a high-RAM runtime')
+        else:
+            print('You are using a high-RAM runtime!')
+
+    # def install_packages(self):
+    #     subprocess.run(["pip", "install", "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git@4e570be9ae4ced8cdc64e498125708e34942befc"])
+    #     subprocess.run(["pip", "install", "--no-deps", "trl<0.9.0", "peft==0.12.0", "accelerate==0.33.0", "bitsandbytes==0.43.3"])
+
+    def prepare_model(self):
+        self.model, self.tokenizer = FastLanguageModel.from_pretrained(
+            model_name=self.config["model_name"],
+            max_seq_length=self.config["max_seq_length"],
+            dtype=None,
+            load_in_4bit=self.config["load_in_4bit"]
+        )
+        self.model = FastLanguageModel.get_peft_model(
+            self.model,
+            r=self.config["lora_r"],
+            target_modules=self.config["lora_target_modules"],
+            lora_alpha=self.config["lora_alpha"],
+            lora_dropout=self.config["lora_dropout"],
+            bias=self.config["lora_bias"],
+            use_gradient_checkpointing=self.config["use_gradient_checkpointing"],
+            random_state=self.config["random_state"],
+            use_rslora=self.config["use_rslora"],
+            loftq_config=self.config["loftq_config"],
+        )
+
+    def process_dataset(self, dataset_info):
+        dataset_name = dataset_info["name"]
+        split_type = dataset_info.get("split_type", "train")
+        processing_func = getattr(self, dataset_info.get("processing_func", "format_prompts"))
+        rename = dataset_info.get("rename", {})
+        filter_data = dataset_info.get("filter_data", False)
+        filter_column_value = dataset_info.get("filter_column_value", "id")
+        filter_value = dataset_info.get("filter_value", "alpaca")
+        num_samples = dataset_info.get("num_samples", 20000)
+
+        dataset = load_dataset(dataset_name, split=split_type)
+
+        if rename:
+            dataset = dataset.rename_columns(rename)
+        if filter_data:
+            dataset = dataset.filter(lambda example: filter_value in example[filter_column_value]).shuffle(seed=42).select(range(num_samples))
+        dataset = dataset.map(processing_func, batched=True)
+        return dataset
+
+    def format_prompts(self, examples):
+        alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
+
+        ### Instruction:
+        {}
+
+        ### Input:
+        {}
+
+        ### Response:
+        {}"""
+        texts = [alpaca_prompt.format(ins, inp, out) + self.tokenizer.eos_token for ins, inp, out in zip(examples["instruction"], examples["input"], examples["output"])]
+        return {"text": texts}
+
+    def load_datasets(self):
+        datasets = []
+        for dataset_info in self.config["dataset"]:
+            datasets.append(self.process_dataset(dataset_info))
+        return concatenate_datasets(datasets)
+
+    def train_model(self):
+        dataset = self.load_datasets()
+        trainer = SFTTrainer(
+            model=self.model,
+            tokenizer=self.tokenizer,
+            train_dataset=dataset,
+            dataset_text_field=self.config["dataset_text_field"],
+            max_seq_length=self.config["max_seq_length"],
+            dataset_num_proc=self.config["dataset_num_proc"],
+            packing=self.config["packing"],
+            args=TrainingArguments(
+                per_device_train_batch_size=self.config["per_device_train_batch_size"],
+                gradient_accumulation_steps=self.config["gradient_accumulation_steps"],
+                warmup_steps=self.config["warmup_steps"],
+                num_train_epochs=self.config["num_train_epochs"],
+                max_steps=self.config["max_steps"],
+                learning_rate=self.config["learning_rate"],
+                fp16=not is_bfloat16_supported(),
+                bf16=is_bfloat16_supported(),
+                logging_steps=self.config["logging_steps"],
+                optim=self.config["optim"],
+                weight_decay=self.config["weight_decay"],
+                lr_scheduler_type=self.config["lr_scheduler_type"],
+                seed=self.config["seed"],
+                output_dir=self.config["output_dir"],
+            ),
+        )
+        trainer.train()
+
+    def inference(self, instruction, input_text):
+        FastLanguageModel.for_inference(self.model)
+        alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
+
+        ### Instruction:
+        {}
+
+        ### Input:
+        {}
+
+        ### Response:
+        {}"""
+        inputs = self.tokenizer([alpaca_prompt.format(instruction, input_text, "")], return_tensors="pt").to("cuda")
+        outputs = self.model.generate(**inputs, max_new_tokens=64, use_cache=True)
+        print(self.tokenizer.batch_decode(outputs))
+
+    def save_model_merged(self):
+        if os.path.exists(self.config["hf_model_name"]):
+            shutil.rmtree(self.config["hf_model_name"])
+        self.model.push_to_hub_merged(
+            self.config["hf_model_name"],
+            self.tokenizer,
+            save_method="merged_16bit",
+            token=os.getenv('HF_TOKEN')
+        )
+
+    def push_model_gguf(self):
+        self.model.push_to_hub_gguf(
+            self.config["hf_model_name"],
+            self.tokenizer,
+            quantization_method=self.config["quantization_method"],
+            token=os.getenv('HF_TOKEN')
+        )
+
+    def prepare_modelfile_content(self):
+        output_model = self.config["hf_model_name"]
+        return f"""FROM {output_model}/unsloth.Q5_K_M.gguf
+
+TEMPLATE \"\"\"Below are some instructions that describe some tasks. Write responses that appropriately complete each request.{{{{ if .Prompt }}}}
+
+### Instruction:
+{{{{ .Prompt }}}}
+
+{{{{ end }}}}### Response:
+{{{{ .Response }}}}\"\"\"
+
+PARAMETER stop ""
+PARAMETER stop ""
+PARAMETER stop ""
+PARAMETER stop ""
+PARAMETER stop "<|reserved_special_token_"
+"""
+
+    def create_and_push_ollama_model(self):
+        modelfile_content = self.prepare_modelfile_content()
+        with open('Modelfile', 'w') as file:
+            file.write(modelfile_content)
+
+        subprocess.run(["ollama", "serve"])
+        subprocess.run(["ollama", "create", f"{self.config['ollama_model']}:{self.config['model_parameters']}", "-f", "Modelfile"])
+        subprocess.run(["ollama", "push", f"{self.config['ollama_model']}:{self.config['model_parameters']}"])
+
+    def run(self):
+        self.print_system_info()
+        self.check_gpu()
+        self.check_ram()
+        # self.install_packages()
+        self.prepare_model()
+        self.train_model()
+        self.save_model_merged()
+        self.push_model_gguf()
+        self.create_and_push_ollama_model()
+
+
+def main():
+    import argparse
+    parser = argparse.ArgumentParser(description='PraisonAI Training Script')
+    parser.add_argument('command', choices=['train'], help='Command to execute')
+    parser.add_argument('--config', default='config.yaml', help='Path to configuration file')
+    args = parser.parse_args()
+
+    if args.command == 'train':
+        ai = train(config_path=args.config)
+        ai.run()
+
+
+if __name__ == '__main__':
+    main()

{praisonai-0.0.57 → praisonai-0.0.59rc2}/praisonai/ui/context.py

@@ -97,17 +97,36 @@ class ContextGatherer:
         return modified_ignore_patterns
 
     def get_include_paths(self):
+        """
+        Loads include paths from:
+            1. .praisoninclude (includes ONLY files/directories listed)
+            2. .praisoncontext (if .praisoninclude doesn't exist, this is used
+               to include all other relevant files, excluding ignore patterns)
+        """
         include_paths = []
-
-        # 1. Load from .praisoninclude
-        include_file = os.path.join(self.directory, '.praisoninclude')
+        include_all = False  # Flag to indicate if we need to include all files
+ 
+        include_file = os.path.join(self.directory, '.praisoncontext')
         if os.path.exists(include_file):
             with open(include_file, 'r') as f:
                 include_paths.extend(
                     line.strip() for line in f
                     if line.strip() and not line.startswith('#')
                 )
-        return include_paths
+ 
+        # If .praisoncontext doesn't exist, fall back to .praisoninclude
+        # for including all relevant files
+        if not include_paths: 
+            include_file = os.path.join(self.directory, '.praisoninclude')
+            if os.path.exists(include_file):
+                with open(include_file, 'r') as f:
+                    include_paths.extend(
+                        line.strip() for line in f
+                        if line.strip() and not line.startswith('#')
+                    )
+                include_all = True  # Include all files along with specified paths
+ 
+        return include_paths, include_all
 
     def should_ignore(self, file_path):
         """
@@ -130,61 +149,78 @@ class ContextGatherer:
                any(file_path.endswith(ext) for ext in self.relevant_extensions)
 
     def gather_context(self):
-        """Gather context from relevant files, respecting ignore patterns and include paths."""
+        """
+        Gather context from relevant files, respecting ignore patterns
+        and include options from .praisoninclude and .praisoncontext.
+        """
         context = []
         total_files = 0
         processed_files = 0
+        self.include_paths, include_all = self.get_include_paths()
 
-        if not self.include_paths:
-            # No include paths specified, process the entire directory
-            for root, dirs, files in os.walk(self.directory):
-                total_files += len(files)
-                dirs[:] = [d for d in dirs if not self.should_ignore(os.path.join(root, d))]
-                for file in files:
-                    file_path = os.path.join(root, file)
-                    if not self.should_ignore(file_path) and self.is_relevant_file(file_path):
-                        try:
-                            with open(file_path, 'r', encoding='utf-8') as f:
-                                content = f.read()
-                                context.append(f"File: {file_path}\n\n{content}\n\n{'='*50}\n")
-                                self.included_files.append(Path(file_path).relative_to(self.directory))
-                        except Exception as e:
-                            logger.error(f"Error reading {file_path}: {e}")
-                    processed_files += 1
-                    print(f"\rProcessed {processed_files}/{total_files} files", end="", flush=True)
-        else:
-            # Process specified include paths
+        def add_file_content(file_path):
+            """Helper function to add file content to context."""
+            try:
+                with open(file_path, 'r', encoding='utf-8') as f:
+                    content = f.read()
+                    context.append(
+                        f"File: {file_path}\n\n{content}\n\n{'=' * 50}\n"
+                    )
+                    self.included_files.append(
+                        Path(file_path).relative_to(self.directory)
+                    )
+            except Exception as e:
+                logger.error(f"Error reading {file_path}: {e}")
+
+        def process_path(path):
+            """Helper function to process a single path (file or directory)."""
+            nonlocal total_files, processed_files
+            if os.path.isdir(path):
+                for root, dirs, files in os.walk(path):
+                    total_files += len(files)
+                    dirs[:] = [
+                        d
+                        for d in dirs
+                        if not self.should_ignore(os.path.join(root, d))
+                    ]
+                    for file in files:
+                        file_path = os.path.join(root, file)
+                        if not self.should_ignore(file_path) and self.is_relevant_file(file_path):
+                            add_file_content(file_path)
+                        processed_files += 1
+                        print(
+                            f"\rProcessed {processed_files}/{total_files} files",
+                            end="",
+                            flush=True,
+                        )
+            elif os.path.isfile(path) and self.is_relevant_file(path):
+                add_file_content(path)
+                processed_files += 1
+                print(
+                    f"\rProcessed {processed_files}/1 files",
+                    end="",
+                    flush=True,
+                )
+
+        if include_all:
+            # Include ALL relevant files from the entire directory
+            process_path(self.directory)
+            
+            # Include files from .praisoninclude specifically
+            for include_path in self.include_paths:
+                full_path = os.path.join(self.directory, include_path)
+                process_path(full_path)
+        elif self.include_paths:
+            # Include only files specified in .praisoncontext
             for include_path in self.include_paths:
                 full_path = os.path.join(self.directory, include_path)
-                if os.path.isdir(full_path):
-                    for root, dirs, files in os.walk(full_path):
-                        total_files += len(files)
-                        dirs[:] = [d for d in dirs if not self.should_ignore(os.path.join(root, d))]
-                        for file in files:
-                            file_path = os.path.join(root, file)
-                            if not self.should_ignore(file_path) and self.is_relevant_file(file_path):
-                                try:
-                                    with open(file_path, 'r', encoding='utf-8') as f:
-                                        content = f.read()
-                                        context.append(f"File: {file_path}\n\n{content}\n\n{'='*50}\n")
-                                        self.included_files.append(Path(file_path).relative_to(self.directory))
-                                except Exception as e:
-                                    logger.error(f"Error reading {file_path}: {e}")
-                            processed_files += 1
-                            print(f"\rProcessed {processed_files}/{total_files} files", end="", flush=True)
-                elif os.path.isfile(full_path) and self.is_relevant_file(full_path):
-                    try:
-                        with open(full_path, 'r', encoding='utf-8') as f:
-                            content = f.read()
-                            context.append(f"File: {full_path}\n\n{content}\n\n{'='*50}\n")
-                            self.included_files.append(Path(full_path).relative_to(self.directory))
-                    except Exception as e:
-                        logger.error(f"Error reading {full_path}: {e}")
-                    processed_files += 1
-                    print(f"\rProcessed {processed_files}/{total_files} files", end="", flush=True)
+                process_path(full_path)
+        else:
+            # No include options, process the entire directory
+            process_path(self.directory)
 
         print()  # New line after progress indicator
-        return '\n'.join(context)
+        return "\n".join(context)
 
     def count_tokens(self, text):
         """Count tokens using a simple whitespace-based tokenizer."""

{praisonai-0.0.57 → praisonai-0.0.59rc2}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "PraisonAI"
-version = "0.0.57"
+version = "0.0.59rc2"
 description = "PraisonAI application combines AutoGen and CrewAI or similar frameworks into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customization, and efficient human-agent collaboration."
 authors = ["Mervin Praison"]
 license = ""
@@ -20,7 +20,7 @@ pyautogen = ">=0.2.19"
 crewai = ">=0.32.0"
 markdown = ">=3.5"
 praisonai-tools = ">=0.0.7"
-pyparsing = ">=3.0.0" # Added to fix a colab issue. Temp Fix.
+pyparsing = ">=3.0.0"
 chainlit = {version = "^1.1.301", optional = true}
 gradio = {version = ">=4.26.0", optional = true}
 flask = {version = ">=3.0.0", optional = true}
@@ -89,6 +89,8 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry.scripts]
 praisonai = "praisonai.__main__:main"
+setup-conda-env = "setup.setup_conda_env:main"
+post-install = "setup.post_install:main"
 
 [tool.poetry.extras]
 ui = ["chainlit"]
@@ -100,4 +102,14 @@ openai = ["langchain-openai"]
 anthropic = ["langchain-anthropic"]
 cohere = ["langchain-cohere"]
 chat = ["chainlit", "litellm", "aiosqlite", "greenlet"]
-code = ["chainlit", "litellm", "aiosqlite", "greenlet"]
+code = ["chainlit", "litellm", "aiosqlite", "greenlet"]
+train = ["setup-conda-env"]
+
+[tool.poetry-dynamic-versioning]
+enable = true
+vcs = "git"
+style = "semver"
+
+[tool.poetry.build]
+generate-setup-file = false
+script = "setup/post_install.py"

praisonai-0.0.59rc2/setup/post_install.py

@@ -0,0 +1,20 @@
+import subprocess
+import sys
+import os
+
+def main():
+    try:
+        # Get the absolute path of the current file 
+        current_file = os.path.abspath(__file__)
+        
+        # Get the directory of the current file
+        script_dir = os.path.dirname(current_file) 
+        
+        # Construct the path to setup_conda_env.py
+        setup_script = os.path.join(script_dir, 'setup_conda_env.py')
+    except subprocess.CalledProcessError as e:
+        print(f"Error occurred while running the setup script: {e}")
+        sys.exit(1)
+
+if __name__ == "__main__":
+    main()